Resuls from the charged jet finder
1. Algorithm
This finder is based only on the charged tracks which are measured in
the TPC. It is based on a CDF finder as described in Phys. Rev. D 65,
092002 (2002). The algorithms takes a pT ordered list of particles and
starts with the highest pT-particle. This particle together with all other
particles in a radius R=0.7 is defined as a jet. It is not required that
the jet-cone is completly in the acceptance. A detailed description of the
algorithm can be found in Section 5.3.
2. Results
2.1 PYTHIA events with 100 GeV/c hard scattering (status 10/31/02)
PYTHIA at 5.5 TeV was used to generate 50,000 events. Standard settings
+ CKIN(3,100), CKIN(4,100.1), CKIN(7,-1), CKIN(8,1), MSEL(1), MSTP(111,1),
MSTP(151,0). This should produce only events with a hard parton scattering
with pT~100GeV/c in the rapidity range -1<y<1.
The charged jet finder was run on each event, the found jet with the
highest pT was used for further analysis. To get some rough estimate
of the acceptance, only charged particles with -1 < eta < 1
where used.
Figures 1 and 2 show position in eta-phi and the pT of the found (highest
pT) jets. The distribution in eta-phi is basically flat, the pT-distribution
has it's maximum around 50 GeV/c. For the further analysis of jet properties,
the found jets where binned into jet-pT bins, as indicated by the colored
bands in Figure 2. The non-flat pT-distribution causes a weird weightening
of jets in the different bins, the mean is NOT in the middle of the pT-bin!
Fig 1: eta-phi distribution of found jets
Fig 2: pT of found jets
Figure 3 shows the number of charged particles assigned to a jet. Different
colors indicate different jet-pT bins (black: 20<pT<30, 30<pT<40,
40<pT<50, 50<pT<60, 60<pT<70, 70<pT<100), see Figure
2. Figure 4 shows the mean (arithmetic mean) number of charged particles
per jet, the vertical error bars show the RMS of the distribution. The pT
of the jet in figure 4 is the arithmetic mean of the jet-pt bin, please
keep in mind that this wrong/strongly biased.
The shift of the most probably value to higher ncharged values is clearly
seen. When one compares this to the CDF paper, I see fewer charged particles
per jet... I would habe expected to see more, since they have a momentum
cut at pT>0.5 Gev/c, while I don't apply a momentum cut. Reason for this
is unclear... Is this bias from the underlying event (might be different
for 5.5 TeV and 1.8 TeV) or the pT (y) cut in PYTHIA (is a 20 GeV/c jet from
a 100 GeV/c hard scattering the same as a mimimum bias 20 GeV/c jet, probably
not...)??? Would be consistent with the fact that I see approximatly the
"right" number of particles for pt = 40-60 GeV/c jets...
Fig 3: Number of associated charged particles in a jet
Fig 4: Mean number of associated charged particles as function of jet
pT
Figure 5 shows the pT-distribution of the associated particles for the
different jet-pT bins. The jump in the pT-spectra (most clearly visible
for the lowest jet-pT bin 20 GeV/c < pT(jet) < 30 GeV/c (black line))
comes from jets with one associated charged particle. Is a one particle
jet a jet??? The CDF algorithm allows it, see figure 3 in PRD 65, 092002
(2002). The jump might cause part of the shift to lower number of particles
in the jet which was observed in figure 3... We see however the expected
hardening of the spectra with increasing jet pT.
Please note the logarithmic color scale.Figure 6 shows the particle-pT
versus the number of charged particles in the jet. The upper left plot shows
the results from jet-pt bin 1, bin 2 is in the upper middle... One clearly
sees the increase of the mean number of charged particles per jet and the
hardening of the pT-spectra. The one particle jets which causes the jump in
the pT-spectra are visible in the first three bins.
Fig 5: pT-spectra of the associated charged particles in a jet
Fig 6: pT of the associated particles vs number of associated particles
for the different jet-pT bins
Figure 7 and 8 show the jet finder radius dependence of pT and number of
charged particles. To generate these plots, the pT (ncharged) for a given
radius around the bin center was calculated and divided by the total pT (ncharged)
of the jets. This yields 2-dimensional histograms. Here are shown the profile
histograms for all ratios on the radius axis for the different jet-pT bins
(colors same as in figure 2).
One clearly sees the decrease of the ratios with decreasing radius.
The pt-ratio drops different for different jet-pT's, with
higher jet-pt more of the pT is containt in a smaller radius. For the ncharged
ratio such a dependence can not be seen. This is consistent with the CDF results,
see figure 6 from PRD 65 092002 (2002). There seems to be a difference in
the absolut value, but this might be connected to the number of particles
per jet, which was observed to be lower for this study (see figure 3+4 and
comments...)
Fig 7: Ratio between pT in a radius R around the jet center and total
pT of the jet
Fig 8: Ratio between number of particles in a radius R around jet center
and total number of particles in the jet
Conclusions:
- The charged jetfinder seems to work :-)
- The analysis software seems to work :-)
- Results are not unreasonable, but there are differences with the
published CDF reference data... A possible explanation might be bias from
the event generation in PYTHIA. To study this, I started to run PYTHIA mimimum
bias (let's say minimized bias ;-) simulations. To get enough statistics
for the higher pt-bins takes some time, so there are no results yet...
- Another possibility to explain the different results might be a inaccuracy
in the algorithm description: I used so far the pT weighted mean as jet position,
while the CDF results may be obtained with the leading particle position
as jet position. I'll try to investigate the influence on the results...
Thorsten Kollegger
Constantin Loizides
IKF - University of Frankfurt
Last updated: 10/31/2002 16:25 pm EST